. Recently two studies reported a strong association between high bone mass and increased risk of breast cancer. According to the applicant, this result raises new questions in decision-making for HRT and suggests a potential use of bone mass as an indicator of lifetime estrogen exposure for assessing risk of breast cancer. Additional studies in different ethnic groups appear warranted, because there is significant ethnic variation in frequencies, distributions and severity levels of both osteoporosis and breast cancer. Given the fact that the bone mineral density (BMD) is higher, but the rate of breast cancer is lower in Hispanic older women compared with Anglo women, the Hispanic postmenopausal women would be a model to further evaluate the relationship between breast cancer and bone mass. To date, most knowledge of risk factors of osteoporosis and breast cancer is mainly based on results from Anglo women. It is critical to examine those risk factors in the Hispanic population in order to form specific prevention strategies for this population. The Women's Health Initiative (WHI) provides a unique opportunity to undertake nested case-control studies (NCCS) to further examine these issues in different ethnic groups. However, within the WHI there will not be a sufficient number of breast cancer cases to form a NCCS in Hispanic women, both because the number of Hispanic women in the WHI BMD study cohorts is small (expected n=800) and because breast cancer rate is lower in Hispanic women compared with Anglo. The proposed study is a case-control research design. The controls will be chosen from the Hispanic participants of the Arizona WHI observation study group. No additional data will be collected from the controls. The case (n=140) will be newly diagnosed Hispanic postmenopausal breast cancer patients in Arizona (obtained through collaborative oncologists and surgeons). Measurements will include anthropometry, body composition, BMD, peripheral white blood cells and WHI questionnaires. The relationship between BMD and breast cancer will be evaluated using logistic regression analysis.